Study on pulse current auxiliary heating process of submerged arc welding

Abstract

During the welding process, temperature filed distribution of weldment is one of the key factors which influence welding quality. In order to improve the mechanical properties of welding joint, a technology for heating process of weldment using auxiliary pulse current was proposed in this article. Firstly, through metallographic experiment, strength experiment and hardness experiment, it was proved that the auxiliary pulse current not only can refine grain size but also improves the mechanical property of the welding joint. Then the paper used ANSYS and its ANSYS Parametric Design Language parametric design language to simulate the welding process with assistant pulse current, and the temperature field, the pulse current density distribution and the time of effective pulse current in the welded joint were obtained. Quantitative analyses were undertaken on the influence laws of auxiliary pulse current parameters on weld temperature field, auxiliary pulse current density distribution and the time of effective pulse current. A new parameter was defined as time-current density (the time integral of current density) to demonstrate the change rule of auxiliary pulse current density distribution with time. Finally, analyzing the reasons for raising the welding quality. There were two reasons for such phenomenon: one was auxiliary pulse current change the state of heat distribution in the weld seam area, the other was that the impulse oscillation of the auxiliary pulse influences the nucleation process of melting region.

1. Introduction

Researches show that welding temperature is one of the most important factor which influences the quality of the welded seam in the process of welding [1]. The large temperature gradient can lead to coarse columnar crystal and equiaxed grains significantly increase, and makes mechanics performance decrease in the welded joint [2]. The existing research also indicated that the performance of the weld can be improved with the regulation of the temperature field distribution [3].

During the welding process with auxiliary pulse current, the auxiliary pulse current not only can improve the distribution of the welding temperature field but also cause electron migration, which have effects on welding quality [4]. Many researches indicated that the auxiliary electric current can evidently refine the grain size and improve the distributed uniformity of solidified structure [5,6]. Using the Joule effect of the electric current, the aging treatment with electric current has become hotspot of material science. The research indicated that the technology of low-density pulse current ageing treatment can improve the microstructures and mechanical properties of AA2219 aluminum alloy [7]. During the preparation of titanium alloy by equal channel angular pressing, applying an auxiliary current to titanium alloy, the course of recrystallization is speeded up and the phase transition is induced. The elongation of titanium alloy obtained on the above-mentioned conditions is three times more than the conventional annealing [8]. Electric current pulse was applied during the hardening of medium carbon steel, the Joule heat affected the dislocation dynamics and dislocation density which induce residual stress decreases [9]. Al–Li alloy and Al-MMCs materials were used to study on pulse current heating behaviors and thermal forming performance. Research results indicate that the high-intensity pulse current flows through the sheet and generates the tremendous Joule heat. The product produced by the said process can exhibit good shape retention, surface quality, and high geometry accuracy [10]. During Al–Si alloy heat treatment with auxiliary current, the course of coarse band-form eutectic silicon granulation is speeded up [11]. In the study of the sheet metal aging heat treatment with low density auxiliary current, Ma found that the peak current density of the specimen was between 40 and 120 A/cm², the increased temperature of the specimen is under 1 °C [12]. KUANG et al. [13] carried out large strain electrophysical plasticity rolling on AZ31 plate and found that TD deflection weave could be prepared on AZ31 plate in a single pass, which improved the formability of magnesium alloy. QIAN et al. [14] carried out cold rolling and electrophysical plasticity rolling on quenched mild steel with different deformations. The results showed that the strength and hardness of the electrophysically rolled specimens were slightly lower than those of the cold rolled specimens with the same underpressure rate, and the elongation was significantly increased; during plastic deformation, the weaving structure of the electrophysically rolled specimens showed obvious rotation, and the weaving structure transformed into a relatively stable {112}<110>. ZHANG et al. [15] used high-frequency pulsed current to assist the magnesium/aluminum alloy one-step rolling and welding composite plate. The results found that the traditional hot rolling preparation of magnesium/aluminum alloy composite plate large downward pressure rate composite when the coordination of heterogeneous metal deformation is poor and not easy to control, and high-frequency pulsed current auxiliary role can be realized in magnesium/aluminum alloy composite plate composite of the small downward rate composite. Kang K J et al. applied pulsed current to the aging treatment of 7050 alloy cast plates, and compared with conventional heat treatment, pulsed current-assisted aging treatment resulted in an increase in the tensile strength of 7050 alloy plates from 428 to 470 MPa, elongation from 3.3 to 6.5%, and hardness from 164 to 185 HV, while the treatment time was shortened from 16 to 8 h [16]. The treatment time was reduced from 16 h to 8 h [16]. Troitsky O A et al. passed pulsed current during the drawing process of metal wires (copper and steel wires) and found that the structure of the wires was significantly improved with a reduction in tensile strength of 30%–35% and a reduction in resistance of 18%–20% [17]. Zhang X et al. found that there is a “targeting effect” (i.e., due to the fact that the resistance of defects is greater than that of the ideal lattice region, resulting in more Joule heat generated in the defective region) in the process of assisted tensile organization of nickel-based high-temperature alloys by pulsed current, and the use of the “targeting effect” can be used to target the control of local microstructures of anisotropic materials and to achieve the homogeneous distribution of macroscopic mechanical properties [18]. Kumar A et al. applied a pulsed current of 100 A/mm2 for 10 s to a prefabricated V-notched steel plate to study the effect of pulsed current on fatigue cracks, and found that the fatigue cracks achieved complete healing under pulsed current, whereas there was no significant change in the microstructure in the region farther away from the cracks [19]. Ma R et al. performed heat treatment and pulsed current treatment on high strength steel containing irreversible hydrogen damage, and the results showed that the mechanical properties of high strength steel containing irreversible hydrogen damage were effectively restored after pulsed current treatment [20].In this thesis, ×80 was used for studying the effects of assistant current on the mechanical property in welded joint. Considering the electric heat cannot be too high, the pulse current is selected as the assistant current. The research method is as follows: First, the experiments of metallographic observations, tensile strength and micro hardness were made, the experiment results could be used to verify the effectiveness of the method by use assistant pulse current to improve weld quality. Secondly, the welding with assistant pulse current process is simulated by the finite element software. And the transient temperature field, the pulse current density distribution and the time of effective pulse current were achieved. Finally, the influence of electric heat on weld quality is analyzed.

2. Experiment materials and methods

In this study, using ×80 pipe line steel as experimental materials, its chemical components as shown in Table 1. Its original structure consists of acicular ferrite (as shown in Fig. 1a). Experiment equipment (as shown in Fig. 1b) includes: pulse electrical source and submerged arc welding machine.

Table 1.

Chemical compositon of ×80 steel (wt%).

C	Si	Mn	P	S	Nb	V	Ti	Ni	Cr	Cu	Mo	N	B
0.051	0.23	1.86	0.011	0.0009	0.076	0.002	0.012	0.39	0.24	0.19	0.24	0.000035	0.0003

Fig. 1.

Welding experimental: (a) Original structure of ×80 steel; (b) Welding experimental equipment.

The thermal property parameters of ×80 steel are shown in Table 2.

Table 2.

Thermal properties of ×80 steel.

Temperature (°C)	Heat conductivity coefficient (W/m·°C)	Specific heat (J/kg·°C)	Density (kg/m3)	Specific resistance (Ω·m × 10−7)
20	50	460	7820	2.2
250	47	480	7770	4.93
500	40	530	7610	5.95
750	27	675	7550	8.75
1000	30	670	7490	11.2
1500	35	660	7350	12.6
1750	140	780	7300	14

The weldment is 250mm × 150mm × 10 mm X80 steel plate. The weldment with V-groove along the weld and the electrodes connection holes. The auxiliary current source is connected to both sides of the weldment. The specific size of the weldment as show in Fig. 2. The welding parameters as show in Table 2.

Fig. 2.

The shapes and sizes of the weldment.

During the experiment of the welding with pulse current, multiple consecutive current pulses are applied with identical pulse width of 160 μs, identical current frequency of 140 Hz and identical voltage of 75 V. The whole process of welding is accompanied with auxiliary pulse current. In order to reveal the effects of pulse current on weld joints′ microstructure and mechanical properties, metallographic observations, tensile strength and micro hardness experiments were made.

3. Experiment results

3.1. Metallographic observation

The metallographic structure of weld joint was observed with a metallographic microscope. The location of the metallographic observation points as shown in Fig. 3.

Fig. 3.

The location of metallographic observation points.

The actual pulse power generation device used in the experiment uses 4-pole and 4-position band switch to control the opening of 4 parallel circuits, adopts the GTR module with good performance and high power, and triggers the high-power GTR module through the driving current provided by Darlington tube. The measured waveform of the output current of the pulse power supply with ultra-low frequency oscilloscope shows that the pulse current has the characteristics of large pulse current and good waveform, which can better meet the requirements of electroplastic machining and theoretical research on the power supply device. The appearance of the pulse power supply for electroplastic machining is shown in Fig. 4, and the pulse current waveform output by the power supply is shown in Fig. 5.

Fig. 4.

Pulse generator outside view.

Fig. 5.

Wave of Pulse current diagram.

Fig. 6, Fig. 7, Fig. 8, Fig. 9, Fig. 10a shows the metallographic structure without pulse current at points A, C, F, G, and I. Fig. 6, Fig. 7, Fig. 8, Fig. 9, Fig. 10b shows the metallographic structure at points A, C, F, G, and I with pulsed current. The figures showed that there was a significant influence of pulse current on metallographic structure of the coarse-grained heat-affected zone. In the region, the additional pulse current not only causes grain refining, but also makes the lath bainite turned into the granular bainite.

Fig. 6.

Microstructures of the observation point A:

a) without pulse current; (b) with pulse current.

Fig. 7.

Microstructures of the observation point C:

(a) without pulse current; (b) with pulse current.

Fig. 8.

Microstructures of the observation point F:

(a) without pulse current; (b) with pulse current.

Fig. 9.

Microstructures of the observation point G:

(a) without pulse current; (b)with pulse current.

Fig. 10.

Microstructures of the observation point I:

(a) without pulse current; (b) with pulse current.

Comparison found that the pulse current on the C point, that is, the welding of the most obvious effect of the coarse grain area. No pulse current when the organization of the coarse grain area is mainly slate bainite, after the application of pulse current organization into a granular bainite-based organization, and the organization is more uniform, fine; on the weld area also has a certain effect, as can be seen from Fig. 6, Fig. 7, Fig. 8, Fig. 9, Fig. 10, after the application of pulse current at the weld the organization of the weld has been refined, but due to the weld at the organization of the weld is uneven, it is not good to analyze the pulse current The degree of influence on the place; from Fig. 8, Fig. 10 can be seen, pulse current on the organization of the weld fine grain area basically has no effect on the application of pulse current welding and conventional welding of the welded joints at the organization of the size is basically equal.

In conclusion, metallographic analysis reveals that pulsed current affects the welded coarse-grained zone to a greater extent than other regions; pulsed current makes the tissue type of the welded coarse-grained zone change (from lath-like bainite to granular bainite), and the tissue becomes fine and homogeneous.

3.2. Hardness test

Hardness test was made by micro hardness testing machine. The load is 0.5 kgf and the loaded time is 10s. The coordinate system and the direction of measuring pressure point as shown in Fig. 3.

Fig. 11a shows the Vickers hardness curve along the point A，B，C direction. Fig. 11b shows the Vickers hardness curve along the point D，E，F direction. Fig. 11c shows the Vickers hardness curve along the point G，H，I direction.

Fig. 11.

The hardness value of the welding joint: (a) along the point A, B, C direction;

(b) along the point D, E, F direction; (c) along the point G, H, I direction.

As Fig. 11 shows, with the application of the current pulses to the welding process, the hardness value decreased compared to conventional weld. But this change is not obvious.

3.3. Tensile test

The dimensions of the tensile samples as shown in Fig. 12a, the practical appearance of the tensile samples as shown in Fig. 12b. The results of this experiment as shown in Table 3. Table 3 demonstrated that the pulse current can increase the yield strength, tensile strength and ductility of the welded samples, reduce the yield ratio (see Table 4).

Fig. 12.

The samples of tensile test: (a) specimen size; (b) specimen appearance.

Table 3.

Welding parameters.

Power source	Welding voltage(V)	Welding current(A)	Welding speed (m/h)	Wire diameter (mm)	Electrode model
DCEP	30	465	20	4	H08A

Table 4.

The results of tension tests.

Welding method	Yidld strength (MPa)	Tensile strenght (MPa)	Yield ratio (%)	Ductility (%)
Without current	583	636	92	18.2
With current	601	681	85	23.2

4. Simulation study

4.1. Finite element model and process parameters

Using ANSYS and its APDL parametric design language simulate the entire welding process. Get the temperature field, the pulse current density distribution and the time of effective current pulse in the welded joint with identical pulse width of 60 μs, identical current frequency of 140 Hz and identical voltage of 75 V. Simulation model was established according to the size of the actual welding experiment sample, as shown in Fig. 3.

Fig. 13a shows the geometry model of the weldment. Owing to irregular geometry and higher temperature gradients in the weld center, mesh refining was conducted for the key positions around the weld center, as shown in Fig. 13b. In welding process, the welding groove was gradually filled by welding wires. To simulate the whole welding process accurately, the technique of birth and death of element was brought in the finite element model of welding simulation. The finite element model after elements death as shown in Fig. 13c. Welding technology parameters as listed in Table 3. The location of the observation points as shown in Fig. 3.

Fig. 13.

The welding finite element model: (a) The geometry model; (b) The finite element mesh generation; (c) The finite element model after elements death.

4.2. Simulation results

4.2.1. Temperature field

The welding temperature field without auxiliary current as shown in Fig. 14. The welding temperature field with auxiliary current as shown in Fig. 15.

Fig. 14.

Welding temperature field without auxiliary current (°C).

Fig. 15.

Welding temperature field with auxiliary current (°C).

According to the results of welding temperature field, the shape of welding pool is similar to two ellipsoids, so the heat source model is double ellipsoid heat source model.

The addition of the pulse current is equivalent to a discontinuous energy loading at the welding joint, which will inevitably cause changes in the distribution of the welding temperature field at the welding joint. However, the pulse current also includes three parameters (pulse voltage, pulse frequency and pulse width), and the temperature rise caused by different parameter combinations will be different. Therefore, the welding temperature field distribution of different combinations of electrical parameters is simulated respectively in the next section to compare the influence of different electrical parameters on the temperature field in each region of the welding joint.

The location of the temperature observation points as shown in Fig. 3. The maximum temperature change value of observation points were caused by auxiliary pulse current as shown in Table 5. It can be seen that the auxiliary current's existence change the temperature field distribution of the welded joints. Table 5 shows that the temperature rise of the heat-affected zone is higher than that of welding joint. The observation points of C, F, G and I which temperature rise are evident, the microstructure is also refined and homogeneous markedly (Fig. 6, Fig. 7, Fig. 8, Fig. 9, Fig. 10 and Table 5).

Table 5.

The maximum temperature change values of observation points.

Obervation point	A	B	C	D	E	F	G	H	I
Temperature change (°C)	42.3	65.7	155.6	45.3	70.6	163.3	226.2	196.4	179.8

4.2.1.1. The effect of auxiliary pulse voltage on temperature filed

While the pulse frequency and pulse width remain unchanged (pulse frequency: 10 Hz, pulse width: 40 μs), the relationship between auxiliary pulse voltage and temperature filed was observed by setting different auxiliary pulse voltage (50 V, 70 V, 90 V and 110 V). When the welding moved to the middle location of two electrodes, the temperature field cloud image of 50 V auxiliary pulse voltage is shown in Fig. 16a. When the welding moved to the middle location of two electrodes, the temperature field cloud image of 70 V auxiliary pulse voltage is shown in Fig. 16b. When the welding moved to the middle location of two electrodes, the temperature field cloud image of 90 V auxiliary pulse voltage is shown in Fig. 16c, When the welding moved to the middle location of two electrodes, the temperature field cloud image of 100 V auxiliary pulse voltage is shown in Fig. 16d. The welding temperature rises with the increase of auxiliary pulse voltage.

Fig. 16.

The temperature field cloud picture under different auxiliary pulse voltages (°C): (a) U = 50 V; (b) U = 70 V; (c) U = 90 V; (d) U = 110 V.

In order to more clearly show the effect of auxiliary pulse voltages change on temperature field, the change in temperature caused by an increase auxiliary pulse voltages were observed. The location of the temperature observation points as shown in Fig. 17. Selecting four observation directions along observation points (line 1, line 2 line 3 and line 4 orientation). The temperature curve of each point under different auxiliary pulse voltages as show in Fig. 18.

Fig. 17.

The location of the temperature observation points.

Fig. 18.

The temperature change curve of each point under different auxiliary pulse voltages: (a) line 1; (b) line 2; (c) line 3; (d) line 4.

Under the same auxiliary pulse voltages, the amount of temperature change firstly decreased and then increased from point 1 to point 4, as show in Fig. 18a. This is due to from point 1 to point 3 is more and more near to the upper surface, the heat release will be more and more. While the point 4 is close to the electrode, the Joule heat is more evident. Under the same auxiliary pulse voltages, the amount of temperature change gradually enhance from point 5 to point 8, as show in Fig. 18b. This is due to the line 2 is located on the middle of the weldment thickness direction, the temperature rise depend on auxiliary pulse current density. Farther away from weld joint, the lower temperature and the higher auxiliary pulse current density and Joule heat are. Under the same auxiliary pulse voltages, the amount of temperature change of point 9 and 10 are basically equal. The temperature variation of point 11 increase obviously, but the temperature variation of point 12 is lower than point 11, as show in Fig. 18c. This is due to the welding heat of the points 9 and 10 are higher, these places permit less electricity to pass through. The amount of temperature change of points 1, 5 and 9 as show Fig. 18d (U = 50 V, f = 10 Hz, μ_s = 40 μs). From the top surface to the bottom surface, the amount of temperature change is gradual growth. This is because the auxiliary pulse current density is distribution-intensive in weldment lower area.

4.2.1.2. The effect of auxiliary pulse width on temperature filed

While the pulse frequency and pulse voltages remain unchanged (pulse frequency: 10 Hz, pulse voltages: 50 V), the relationship between auxiliary pulse width and temperature filed was observed by setting different auxiliary pulse width (40 μs, 60 μs, 80 μs, and 100 μs). When the welding moved to the middle location of two electrodes, the temperature field cloud image with an auxiliary pulse width of 40 μm is shown in Fig. 19a. When the welding moved to the middle location of two electrodes, the temperature field cloud image with an auxiliary pulse width of 60 μm is shown in Fig. 19b. When the welding moved to the middle location of two electrodes, the temperature field cloud image with an auxiliary pulse width of 80 μm is shown in Fig. 19c. When the welding moved to the middle location of two electrodes, the temperature field cloud image with an auxiliary pulse width of 100 μm is shown in Fig. 19d.

Fig. 19.

The temperature field cloud picture under different auxiliary pulse width (°C): (a) μ_s = 40 μs; (b) μ_s = 60 μs; (c) μ_s = 80 μs; (d) μ_s = 100 μs＀

The welding temperature rises with the increase of auxiliary pulse width. The temperature curve of each point under different auxiliary pulse width as show in Fig. 20. The location of the temperature observation points as shown in Fig. 17.

Fig. 20.

The temperature change curve of each point under different auxiliary pulse width:

(a) line 1; (b) line 2; (c) line 3; (d) line 4.

The auxiliary pulse width increase result in the action time of electric current increase, and the current produces more Joule heat. The temperature of the observation points increased with the increasing of the auxiliary pulse width. Under the same auxiliary pulse width, the amount of temperature change firstly decreased and then increased from point 1 to point 4, as show in Fig. 20a. The amount of temperature change gradually enhance from point 5 to point 8, as show in Fig. 20b. The temperature variation of point 11 increase obviously, but the temperature variation of point 12 is lower than point 11, as show in Fig. 20c. The amount of temperature change of point 1, 5 and 9 as show Fig. 20d (U = 50 V, f = 10 Hz, μ_s = 60 μs). From the top surface to the bottom surface, the amount of temperature change is gradual growth, but the increasing range deduce.

4.2.1.3. The effect of auxiliary pulse frequency on temperature filed

While the pulse width and pulse voltages remain unchanged (pulse width: 40 μs, pulse voltages: 50 V), the relationship between auxiliary pulse frequency and temperature filed was observed by setting different auxiliary pulse frequency (10 Hz, 20 Hz, 40 Hz and 50 Hz). When the welding moved to the middle location of two electrodes, the cloud image of temperature field with auxiliary pulse frequency of 10 Hz is shown in Fig. 21a. When the welding moved to the middle location of two electrodes, the cloud image of temperature field with auxiliary pulse frequency of 20 Hz is shown in Fig. 21b. When the welding moved to the middle location of two electrodes, the cloud image of temperature field with auxiliary pulse frequency of 40 Hz is shown in Fig. 21c. When the welding moved to the middle location of two electrodes, the cloud image of temperature field with auxiliary pulse frequency of 50 Hz is shown in Fig. 21d. The welding temperature rises with the increase of auxiliary pulse frequency. The temperature curve of each point under different auxiliary pulse frequency as show in Fig. 22. The location of the temperature observation points as shown in Fig. 17.

Fig. 21.

The temperature field cloud picture under different auxiliary pulse frequency (°C): (a) f = 10 Hz; (b) f = 20 Hz; (c) f = 40 Hz; (d) f = 50 Hz.

Fig. 22.

The temperature change curve of each point under different auxiliary pulse frequency:

(a) line 1; (b) line 2; (c) line 3; (d) line 4.

The pulse frequency curves of observation points 1, 2, 3 and 4 with temperature are shown in Fig. 22a. The pulse frequency curves of observation points 5, 6, 7 and 8 with temperature are shown in Fig. 22b. The pulse frequency curves of observation points 9, 10, 11 and 12 with temperature are shown in Fig. 22c. The amount of temperature change of points 1, 5 and 9 as show Fig. 22d (U = 50 V, f = 20 Hz, μ_s = 40 μs). From the top surface to the bottom surface, the amount of temperature change is gradual reduced.

The above simulation results indicate that the amount of temperature change is basically determined by auxiliary pulse current distribution and heat-transfer conditions. Through contrasting the amount of temperature change caused by auxiliary pulse current, it can be showed that the amount of temperature change of the points 4, 7 and 8 are most obvious. The points 4, 7 and 8 are located in welding heat affected zone, as a conclusion, the auxiliary pulse current have the greatest effects on the welding heat affected zone.

4.2.2. Auxiliary pulse current density distribution

Experimental results show that the metal structure in high temperature can be remarkably refined by the application of the higher density pulsed current. This phenomenon could be related closely to the pulse current density. In order to study the distribution of pulse current density in welding area, the potential distribution cloud picture and current density distribution vector graph are obtained by numerical simulation (U = 50 V, f = 20 Hz, μ_s = 40 μs), as show in Fig. 23, Fig. 24 respectively.

Fig. 23.

The potential distributions cloud picture.

Fig. 24.

The current density distribution vector graph.

When the welding proceeds to the position between tow electrodes of auxiliary pulse current, the auxiliary pulse current density values were obtained from the observation points separately, as shown in Table 6. The location of the observation points as shown in Fig. 3. It is clearly observed that the electric current density distribution in welding area are different. This is caused by the uneven distribution of welding temperature.

Table 6.

The electric current density values of observation points.

Obervation point	A	B	C	D	E	F	G	H	I
Electric current density ( × 108 A/m2)	3.21	3.75	5.61	3.49	4.63	5.68	7.67	6.76	5.77

4.2.2.1. The effect of auxiliary pulse voltages on auxiliary pulse current density distribution

While the pulse width and pulse frequency remain unchanged (pulse width: 40 μs, pulse frequency: 10 Hz), the relationship between auxiliary pulse voltages and auxiliary pulse current density distribution was observed by setting different auxiliary pulse voltages (50 V, 70 V, 90 V and 110 V).

When the welding moved to the middle location of two electrodes, the auxiliary pulse current density at each observation points under different auxiliary pulse voltages as shown in Fig. 25. The location of the observation points as shown in Fig. 17. Fig. 25 indicates that the auxiliary pulse current density at every point increased as the auxiliary pulse voltages increasing. At point 9, the auxiliary pulse current density is bigger than the other point. This is due to the uneven distribution of temperature cause flow around phenomenon.

Fig. 25.

The auxiliary pulse current density under different auxiliary pulse voltages.

4.2.2.2. The effect of auxiliary pulse width on auxiliary pulse current density distribution

While the pulse voltages and pulse frequency remain unchanged (pulse voltages: 50 V, pulse frequency: 10 Hz), the relationship between auxiliary pulse width and auxiliary pulse current density distribution was observed by setting different auxiliary pulse width (40 μs, 60 μs, 80μs and 100 μs). When the welding moved to the middle location of two electrodes, the auxiliary pulse current density at each observation points under different auxiliary pulse width as shown in Fig. 26. The location of the observation points as shown in Fig. 17. The function time of current increases with the increase of auxiliary pulse width. This led to the temperature and resistance increases, which led to a reduction in the auxiliary pulse current density.

Fig. 26.

The auxiliary pulse current density under different auxiliary pulse width.

4.2.2.3. The effect of auxiliary pulse frequency on auxiliary pulse density distribution

While the pulse voltages and pulse width remain unchanged (pulse voltages: 50 V, pulse width: 40 μs), the relationship between auxiliary pulse width and auxiliary pulse current density distribution was observed by setting different auxiliary pulse frequency (10 Hz, 20 Hz, 40 Hz, and 50 Hz). When the welding moved to the middle location of two electrodes, the auxiliary pulse current density at each observation points under different auxiliary pulse frequency as shown in Fig. 27. The location of the observation points as shown in Fig. 17. Like the pulse width, the function time of current increases with the increase of auxiliary pulse frequency. This led to the temperature and resistance increases, which led to a reduction in the auxiliary pulse current density.

Fig. 27.

The auxiliary pulse current density under different auxiliary pulse current frequency.

4.2.3. Effective current function time

The existing studies have shown that pulse current can urge the liquid and high temperature solid metal organization to change [21]. The affects not only depends on the parameters of electrical current, but also on the function time of electrical current. For this paper we define the effective current as the current passing through the finite element with temperature reach 750 °C, since the phase transition temperature of ×80 pipe line steel is 750 °C. The effective current function time of observation points as shown in Table 7. The location of the observation points as shown in Fig. 3。

Table 7.

The effective current function time values of observation points.

Obervation point	A	B	C	D	E	F	G	H	I
Time(s)	6.42	6.00	4.74	6.36	6.00	5.04	6.30	6.00	4.92

In order to analyze the relation between the effective current function times and weld properties, the settings of time stamp are as follows: t₁ is the time at temperatures between 750 and 1540° during the observation points heating stage; t₂ is the time at temperatures above 1540° during the observation points heating stage; t₃ is the time at temperatures above 1540° during the observation points cooling stage; t₄ is the time at temperatures between 750 and 1540° during the observation points cooling stage; t′ is the effective current function times of the observation points; $t_1^{\prime }$, $t_2^{\prime }$, $t_3^{\prime }$ and $t_4^{\prime }$ is the effective current function times of the observation points during t₁, t₂, t₃ and t₄; n is the effective pulse number. The relationship between the above parameters could be expressed as:

$$t_i^{\prime }=t_i\cdot f\cdot {\mu }_{\mathrm{s}}(i=1,2,3,4)$$ (1)

$$t^{\prime }=\sum _{i=1}^4{t}_i(i=1,2,3,4)$$ (2)

$$n=t_i\cdot f(i=1,2,3,4)$$ (3)

where t_i is the effective current function times, f is auxiliary pulse current frequency, μ_s is the auxiliary pulse current width.

The effective current function times of the observation points as shown in Fig. 28. The effective pulse number of the observation points as shown in Fig. 29.

Fig. 28.

The effective current function times.

Fig. 29.

The effective pulse number.

As can be seen from Fig. 28, Fig. 29, the effective current function times in heating stage is much less than cooling stage, the effective current function times of liquid metal is much less than high temperature solid metal. The effective current function times is approx. at 10000μs–15000 μs, so the auxiliary pulse current have enough time to refine microstructure.

4.2.4. Time-current density (the time integral of current density)

Through the above analysis, there is discovery that the auxiliary pulse current density change over time. In order to analyze the variation rules of the auxiliary pulse current density in welding area, a new parameter of time-current density S (the time integral of current density) is established. S₀ is the time-current density over the effective current function time, the time-current density S₁, S₂, S₃ and S₄ are over the effective current function time $t_1^{\prime }$, $t_2^{\prime }$, $t_3^{\prime }$ and $t_4^{\prime }$, respectively. The relationship between the above parameters could be expressed as:

$$S_i=\int J_{i}d{t}_i^{\prime }(i=1,2,3,4)$$ (4)

$$S_0=\sum _{i=1}^4{S}_i(i=1,2,3,4)$$ (5)

where J_i is auxiliary pulse current density at different time, $t_i^{\prime }$ is the effective current function time at different stages.

While the pulse width, pulse density and pulse frequency remain unchanged (pulse width 40 μs, pulse voltage 20 V and pulse frequency 10 Hz), the time-current density of the observation points as shown in Table 8. As seen from the table, the time-current density during the effective current function time $t_4^{\prime }$ is the biggest. Besides, the time-current density at the bottom of the weld joint is larger than other places. The location of the time-current density observation points as shown in Fig. 17.

Table 8.

The time-current density of the observation points.

Time-current density ( × 106 s A/m2)	1	2	3	4	5	6	7	8	9	10	11	12
S1	0.032	0.029	0.043	0.048	0.037	0.044	0.037	0.075	0.169	0.055	0.076	0.151
S2	0.091	0.111	0.039	0.026	0.136	0.106	0.050	0.000	0.227	0.073	0.018	0.000
S3	0.292	0.329	0.211	0.042	0.362	0.308	0.136	0.000	0.718	0.196	0.018	0.000
S4	2.299	2.839	2.538	2.233	3.050	2.778	2.456	1.974	7.225	2.466	2.265	1.896
S0	2.714	3.308	2.831	2.349	3.586	3.236	2.678	2.049	8.340	2.790	2.377	2.047

5. Analysis and discussion

The corresponding relations between temperature field and microstructure of welding zone were validated by comparing simulation results with microstructure observation. The results of comparison showed that the grains were refined in the position of larger Joule heat, the higher its thermal conductivity and specific heat capacity.

The additional pulse current can have an impact on microstructures and properties in welding joint during submerged arc welding. This is mainly because the highly non-uniform welding temperature file lead to the pulse electrical parameters vary from point to point during the welding with pulse current. It differs not only in effective current function time but also in current density. Hence there are various contraction effects at different locations, which will form the contractile force gradient in welding joint. Dendrite broken occurs when the contraction force enough. The broken fragments of the dendrite can act as nucleation particles, and promote heterogeneous nucleation process, eventually making weld microstructure refining.

Metallographic observations show that the additional pulse current has a greater effect on welding heat affected zone. Simulation results show that the additional pulse current produce large temperature rise and current density on welding heat affected zone, and the shortest effective current function time. Considering the pulse electrical parameters are smaller, the electromagnetic effects are not evident. The weldment temperature change caused by the additional electric current should be investigated as a cause for the weld mechanical properties change.

6. Conclusions

The existing studies have shown that pulse current can cause the change of the liquid and high temperature solid metal organization. This is due to when pulse current through the high temperature metal, a pulse magnetic field is produced. The convergent force and shock wave are created by the interaction between the pulse current and pulse magnetic field, and they are the major reasons for the change of solidification microstructure [22,23]. Now, the application of electric current pulse in metal solidification has increasingly becomes new type of technique for microstructure refinement of metals.

Other than the influence factor has been mentioned above, the accession of additional pulse current also has balanced weldment temperatures effect in the welding process. This is due to the most additional current flow through the low temperature zone of the weldment. The temperature rise in low temperature zone caused by the electric heat is more evident. The welding quality would be improved with the regulation of the temperature field distribution.

In this paper, the welding characteristic of ×80 pipe line steel by submerged arc welding with assistant pulse current was investigated. The influence of the auxiliary pulse current parameter on welding temperature field, auxiliary pulse current density distribution and the effective current function time were analyzed. Beside, time-current density was defined, for analyzing the variation rules of the auxiliary pulse current density in welding area. The research contents and results were listed as follows:

• (1)The auxiliary pulse current can change the temperature distribution and deduce the temperature difference of the weld seam.
• (2)The numerical simulation and experimental result indicated that the auxiliary pulse current has a larger influence on welding heat affected area, which obviously improves the mechanical property of welded joints.
• (3)With the temperature of 750 °C–1540 °C during welded joint cooling stage, the time-current density was obviously higher than that of the other stage. This indicates the auxiliary pulse current had a greater role in weld properties during the cooling stage.

The results of this article not only lay the foundation for further study on the technology for heating process of weldment using auxiliary pulse current, but also has the reference value to the study on the auxiliary pulse current can improves the mechanical property of the welding joint.

CRediT authorship contribution statement

Da-long Li: Investigation, Methodology. Xiang-yu Sun: Writing – original draft, Writing – review & editing, Writing – review & editing. Ming-ming Shen: Formal analysis, Validation. En-lin Yu: Data curation, Resources, Software. Jing-chun Niu: Formal analysis, Resources, Software.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.